CN1930588A - Adaptive sampling along edges for surface rendering - Google Patents
Adaptive sampling along edges for surface rendering Download PDFInfo
- Publication number
- CN1930588A CN1930588A CNA2005800079332A CN200580007933A CN1930588A CN 1930588 A CN1930588 A CN 1930588A CN A2005800079332 A CNA2005800079332 A CN A2005800079332A CN 200580007933 A CN200580007933 A CN 200580007933A CN 1930588 A CN1930588 A CN 1930588A
- Authority
- CN
- China
- Prior art keywords
- sampling
- frequency signal
- carry out
- sampled
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T15/00—3D [Three Dimensional] image rendering
- G06T15/50—Lighting effects
- G06T15/503—Blending, e.g. for anti-aliasing
Landscapes
- Engineering & Computer Science (AREA)
- Computer Graphics (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
- Image Processing (AREA)
- Analysing Materials By The Use Of Radiation (AREA)
- Image Generation (AREA)
Abstract
High frequency signals cannot be reconstructed properly from sampled data if the sampling frequency lies below the Nyquist rate. The invention addresses this problem by choosing few additional sample points along a trajectory intersecting the region comprising the high frequency signals, such as an edge. Intermediate rendering data is used to determine the additional sample points. Therefore, according to an exemplary embodiment of the present invention, 4 adaptively chosen sample points per pixel may provide a visual quality comparable to 16 times super-sampling, but at a much lower computational cost.
Description
The present invention relates to the digital imagery field.Especially, the present invention relates to the structure in the data set adaptively sampled method, be used for the structure in the data set is carried out adaptively sampled image processing equipment, scanner system and computer program.
Endoscopy is a kind of important techniques in the medical diagnosis.Usually have only the symptom origin by just may determining disease to the visual inspection of living person's body interior or the sign of commitment.There is every year thousands of endoscopic procedure to be implemented.The endoscope optical inspection is to be used to check the hollow organ of inside of human body or the medical procedure of body cavity.This is a kind of operation of misery normally, wherein by natural duct or by little otch endoscope is inserted in patient's body.Endoscopy is different according to being examined the difference of organ.Wherein a kind of topmost endoscopy is colonoscopy.
Endoscopic another example is a bronchoscopy, and it can be checked patient's lung.Colonoscopy and bronchoscopy all may make the patient incur danger and be uncomfortable.
Virtual endoscopy is the newer diagnostic techniques of a kind of utilization to the Computer Processing of three-dimensional image data set.These data sets provide with the images category that produces by the standard endoscope scrutiny program like or the simulation to patient's organ of equivalence visual.These data sets are to adopt tomography (such as computer tomography (CT) or magnetic resonance imaging (MRI)) to create.Therefore, data acquisition is noninvasive, this means the discomfort that only can cause minimum level to the patient.
Traditional CT and MRI scanning provide the xsect or the sectioning image of health.Each data slicer is the density value that equidistantly distributes that disperses, promptly so-called voxel (voxel).This discretize may produce sampling and reconstruction error.Therefore, the xsect that draws from tomography never the perfection of human body section represent.
Can adopt complicated algorithm and high-performance calculation that a large amount of this xsects are provided, with as direct three dimensional representation to human anatomic structure.The visual danger of having avoided true endoscopy to bring of virtual endoscopy, and if carrying out using virtual endoscopy before the actual endoscopy, the difficulty of this program is minimized.In addition, virtual endoscopy can be checked the maccessiable body region of true endoscopy, such as plat vessels.
Isodensity surface rendering (iso-surface rendering) is used in the visual many clinical practices of three dimensional medical data.The isodensity surface is the set of being had a few that is endowed same density value in the data volume.This data value is called as threshold value or isodensity value (iso-value), and must be specified before checking processing by the operator.Its mark the higher density zone that can be called as the isodensity interior surface with can be called as outside, isodensity surface than the edge between the low-density.The each point that can accurately be mapped in this volume of voxel of this data set is endowed this voxel value.Interpolation around utilizing between the density value of voxel is calculated the density value of other point.By isodensity surface algorithm density value is mapped to inside greater than the point of this threshold value, every other point is mapped to the outside.
An example at the isodensity surface rendering that carries out during the above-mentioned virtual endoscopy carries out during the virtual coloscope inspection.In virtual coloscope is checked, play up the pseudo-realistic views of colon by the CT data.Colon is passed in generation along precalculated path film is common application during virtual coloscope is checked.The generation in this path is semi-automatic or automatically carries out, thereby the clinician can be primarily focused on during so-called " (fly-through) passed through in flight " on the searching polyp.If find this anatomical abnormalities, it is useful leaving the path that generates in advance so and observing this polyp from any viewpoint.Therefore, interactive mode is played up the non-interactive type film generation that helps and can finally replace in the virtual coloscope inspection.
Usually, condition precedent and main standard that surface rendering is used are high-caliber image qualities, because visible artifacts can be disperseed clinician's notice, perhaps even the judgement that leads to errors.
Another important demand is a speed.In order to challenge traditional endoscopy, virtual endoscopy system should have the ability of carrying out interactive navigation in the human organ of being checked.
Regrettably, the design of virtual endoscopy program or method is usually included between these two crucial requirements and trades off, and is very difficult because Gao Gengxin is provided on the one hand and plays up speed and high image quality is provided on the other hand.
A shortcoming being played up the speed generation by height is an aliasing.Aliasing causes along the staircase artifacts of edge and institute's rendering image.This can cause image quality decrease, when with high frame rate render view or when generating film, picture quality can worsen more, for example just has this situation in many virtual endoscopies are used.Because pixel value switches between different gray-scale values at once when the edge exceeds the limit of pixel center, the edge that therefore passes image pixel produces flicker.This typical aliasing effect is very horrible and should eliminate, thereby reaches the target of excellent picture quality.
The over-extraction sample is a kind of method of removing the aliasing pseudomorphism.Here, not to adopt only sample point of each pixel, but assess 16 or multisample and give this pixel more mean value.Because pixel value gradually changes between gray level and passes the edge, thereby can obtain level and smooth edge, and the flicker in the film has also disappeared.Yet this method assesses the cost higher and to play up the target of performance inconsistent with height.Even in the time only need carrying out the over-extraction sample to the several pixels in the image, it also is unacceptable playing up slowing down of speed.
An object of the present invention is to provide a kind of adaptively sampledly, play up performance thereby improve.
According to one embodiment of the present of invention as claimed in claim 1, above-mentioned purpose can be carried out adaptively sampled method to the structure in the data set and realizes by a kind of, wherein this method may further comprise the steps: discern the zone that comprises high-frequency signal of this structure and carry out first sampling with first sampling rate, thereby draw the structure that is sampled that comprises each first sample point.Thereby carry out first sampling along traversing this regional track that comprises high-frequency signal.
Advantageously, according to this one exemplary embodiment of the present invention, select several additional sample points along the track that traverses the zone that comprises high-frequency signal.The orientation of this track can be for perpendicular to this regional direction that comprises high-frequency signal.By selecting first sampling rate enough high, can correctly rebuild high-frequency signal from the data of being sampled, because sample frequency is on the Nyquist rate.Advantageously,, compare, can reduce the sample point number of each pixel, thereby improve sample rate with the over-extraction sample by carrying out first sampling (it is an one dimension) along this track rather than carrying out sampling along the both direction that limits a zone.
According to another one exemplary embodiment of the present invention as claimed in claim 2, discern the structural region that comprises high-frequency signal by carry out second sampling with second sampling rate, wherein first sampling rate is higher than second sampling rate.
Therefore, needn't be to total but only the structural region that comprises high-frequency signal is carried out sampling with the first higher sampling rate.After this structure being sampled and after the identification region of high frequency signals with the second lower sampling rate, can sample to this region of high frequency signals with first (higher) sampling rate, thereby when the quick sampling speed suitable with common over-extraction sample still is provided, reduce the aliasing pseudomorphism in this specific region.
According to another one exemplary embodiment of the present invention as claimed in claim 3, play up the structure of being sampled based on first sampling, wherein determine to comprise the average pixel value of the pixel in this zone of high-frequency signal by the mean value of assessing the first sample point value.
Advantageously, according to this one exemplary embodiment of the present invention, the value along first sample point of the track setting of traversing the zone that comprises higher frequency signals is asked on average, and this mean value is given the specific pixel in this zone subsequently.Can reduce the aliasing pseudomorphism in this region of high frequency signals like this.
According to another one exemplary embodiment of the present invention as claimed in claim 4, determine described structure by isodensity surface rendering program.
Advantageously, these density surface rendering program can provide the quick identification to this structure (for example internal).Advantageously, these density surface rendering program even interactive threshold value adjustment can be provided.
According to another one exemplary embodiment of the present invention as claimed in claim 5, be a line that traverses the zone that comprises high-frequency signal along its track that carries out first sampling.
Advantageously, can further improve sample rate by carry out first sampling along a line.
According to another one exemplary embodiment of the present invention as claimed in claim 6, comprise that the described structural region of high-frequency signal is the edge.Advantageously, according to this one exemplary embodiment of the present invention, can reduce or compensate the aliasing pseudomorphism that occurs along the edge adaptively.
According to another one exemplary embodiment of the present invention as claimed in claim 7, the described adaptively sampled ray that comprises throws (ray casting), and the projection of described ray is used to detect the edge.According to this one exemplary embodiment of the present invention, the ray projection is used to generate the image of this structure that for example has low resolution.Then, according to an aspect of the present invention, identification comprises the image area of this structure of high-frequency signal, and improves sample frequency along traversing this regional track.
According to another one exemplary embodiment of the present invention as claimed in claim 8, described structure comprises the surface, and this surface comprises that the surface changes.In addition, this track extends along the surperficial change direction of maximum.
Advantageously, according to this one exemplary embodiment of the present invention, the direction of change in orientation maximum is carried out first sampling (having upper frequency) surfacewise, and it for example can be the edge.Therefore, according to this one exemplary embodiment of the present invention, on the direction of surface orientation variation maximum, provide additional sample point, so the variation of these sample point values is maximum.
According to another one exemplary embodiment of the present invention as claimed in claim 9, this method is used for virtual endoscopy.This can provide adaptively sampled in the medical applications, for example virtual coloscope inspection, thus the performance of playing up of improvement is provided.
Advantageously, according to another one exemplary embodiment of the present invention as claimed in claim 10, provide a kind of being used for that the structure in the data set is carried out adaptively sampled image processing equipment, it comprises the image processor that is used to store the storer of this data set and is suitable for carrying out following operation: load this data set, identification comprises the structural region of high-frequency signal and carries out first sampling with first sampling rate, thereby draws the structure that is sampled that comprises each first sample point.Carry out first sampling along traversing this regional track that comprises high-frequency signal.
Advantageously, can carry out so adaptively sampled.
The invention still further relates to scanner system, described scanner system comprises the storer that is used for stored data sets and is suitable for the structure in this data set is carried out adaptively sampled image processor.According to this aspect of the present invention, this scanner system is the wherein a kind of of CT scan device system and MR scanner system.This scanner system according to the present invention is illustrated in claim 11 and 12.
Advantageously, can carry out adaptively sampled like this and can obtain by CT scan device system or MR scanner system the performance of playing up to the improvement of the structure in the data set of being gathered.
The invention still further relates to a kind of computer program, it can for example be gone up at processor (such as image processor) and carry out.This computer program can be the part of CT scan device system or MR scanner system for example.According to another one exemplary embodiment of the present invention, these computer programs are illustrated in claim 13.These computer programs can preferably be loaded in the working storage of image processor.Thereby so the image processor of assembling can be realized the one exemplary embodiment of method of the present invention.Described computer program can be stored on the computer-readable medium, such as CD-ROM.These computer programs also can provide (such as WWW) on network, and can the working storage from these network download to image processor.Computer program according to this one exemplary embodiment of the present invention can be write as with any suitable language, for example C++.
As the main points of one exemplary embodiment of the present invention, can see, the structure in the data set is carried out adaptively sampled, wherein carry out sampling with higher sample rate along the track that traverses the zone that comprises high-frequency signal.Therefore, according to an one exemplary embodiment of the present invention, can sample to the major part of this structure with low sampling rate, and only can cause the region of high frequency signals of aliasing pseudomorphism partly to be sampled with higher sample rate to those.Advantageously,, carry out sampling with higher sample rate, can obtain having the suitable rendering image visual quality of over-extraction sample of much more sample like this with every pixel along line perpendicular to region of high frequency signals according to an one exemplary embodiment of the present invention.Assess the cost thereby can cause when sufficiently high picture quality still is provided, reducing.
The embodiment that these and other aspect of the present invention will be described hereafter becomes more obvious, and with reference to the embodiment that after this describes the present invention is set forth.
One exemplary embodiment of the present invention is described below with reference to accompanying drawings:
Fig. 1 illustrates the simplicity of illustration according to an embodiment of computer tomography of the present invention (CT) scanner.
Fig. 2 illustrates the simplicity of illustration according to an embodiment of magnetic resonance of the present invention (MR) scanner.
Fig. 3 illustrates original image and with normal sampling, with comprising the anti-aliasing of over-extraction sample and using diagram according to the anti-aliasing image of playing up of the self-adaptation that comprises the over-extraction sample of one exemplary embodiment of the present invention.
The image that Fig. 4 illustrates usefulness (right side) and need not (left side) checks according to the anti-aliasing virtual coloscope of playing up of the self-adaptation that comprises the over-extraction sample of one exemplary embodiment of the present invention.
Fig. 5 illustrates the image of playing up with normal sampling that demonstrates staircase artifacts (left side), with the anti-aliasing image of playing up that comprises the over-extraction sample (in), and use according to the anti-aliasing image of playing up of the self-adaptation that comprises the over-extraction sample of one exemplary embodiment of the present invention (right side).
Fig. 6 illustrates the process flow diagram that the structure in the data set is carried out an one exemplary embodiment of adaptively sampled method according to of the present invention.
Fig. 7 illustrates the one exemplary embodiment of the present invention according to image processing equipment of the present invention of an one exemplary embodiment that is used to carry out the method according to this invention.
Fig. 1 illustrates the one exemplary embodiment of the present invention according to CT of the present invention (computer tomography) scanner system.With reference to this one exemplary embodiment, the present invention will be described for imaging of medical.Yet, should be noted that to the invention is not restricted to the imaging of medical Application for Field, but also can be used for such as the baggage check in the baggage item to detect as in the application of the objectionable impurities of explosive one class or other commercial Application (such as testing of materials).
Scanner shown in Fig. 1 is the conical beam CT scanner.CT scan device shown in Fig. 1 comprises around the stand 1 of turning axle 2 rotations.This stand drives by motor 3.Reference numeral 4 indication radiation sources (such as x-ray source), it launches the polychromatic radiation bundle according to an aspect of the present invention.
Pencil-beam 6 is guided to the objects 7 that penetrates the center (being in the inspection area of CT scan device system) that is placed on stand 1 and strikes on the detecting device 8.As shown in Figure 1, detecting device 8 is set at the position relative with radiation source 4 on the stand 1, thereby the surface of detecting device 8 is covered by pencil-beam 6.Detecting device 8 shown in Fig. 1 comprises a plurality of detector element.
In the scan period to objects 7, radiation source 4, aperture system 5 and detecting device 8 rotate on arrow 16 indicated directions along stand 1.For making stand 1 rotation that has radiation source 4, aperture system 5 and detecting device 8, motor 3 is connected to motor control unit 17, and Motor Control single 17 is connected to computing unit 18.
In Fig. 1, objects is placed on the travelling belt 19.In the scan period to objects 7, when stand 1 rotated around patient 7, travelling belt 19 made objects 7 move along the direction of the turning axle 2 that is parallel to stand 1.Thus, along helical scan path this objects 7 is scanned.Travelling belt 19 also can stop in scan period.Replace being provided with travelling belt 19, for example in to the medical applications that likes the patient, can adopt removable bed.Yet should be noted in the discussion above that and can carry out circular scan yet in described whole situations, do not have displacement this moment on the direction that is parallel to turning axle 2, but the rotation of stand 1 around turning axle 2 only arranged.
Detecting device 8 is connected to computing unit 18.Computing unit 18 receives testing result (i.e. the reading of the detector element of self-detector 8), and determines scanning result based on these readings.Each detector element of detecting device 8 can be suitable for measuring the decay that conical beam is produced by objects.In addition, computing unit 18 is communicated by letter with motor control unit 17, so that the motion of stand 1 is coordinated mutually with motor 3 and 20 or coordinated mutually with travelling belt 19.
Can realize described computing unit with the data processor that is integrated in the image processing equipment that comprises the storer that is used for stored data sets, this computing unit can also be suitable for carrying out adaptively sampled to the structure in this data set.Data processor according to an aspect of the present invention or image processor can be suitable for loading this data set, and are suitable for discerning the zone that comprises high-frequency signal of this structure.In addition, described data processor can be suitable for carrying out first sampling with first sampling rate, thereby draws the structure that is sampled that comprises each first sample point, wherein carries out first sampling along traversing this regional track that comprises high-frequency signal.
In addition, as shown in Figure 1, computing unit 18 can be connected to loudspeaker 21, with for example automatic output alarm.
Fig. 2 illustrates the simplicity of illustration according to an embodiment of MR scanner system of the present invention.The MR scanner system comprises coil 210, and this coil 210 is provided with and centers on along axle 218 checks that space 217, the patient 215 of examine are positioned at inspection space 217.Advantageously, this patient lies on removable bed or the travelling belt 216, and this removable bed or travelling belt are placed on the bottom of checking space 217.Coil system 210 around this inspection space 217 comprises HF coil 219, comprises the gradient coil 213 and the active shielding coil of Inside coil or shields 212 active shielding arrangement and cryostat 211, during magnetic field produced, these coils were set in this cryostat so that be cooled.The arrangement of gradient coil 213,212 can be connected to gradient amplifier 220.
In addition, according to CT scan device system as shown in Figure 1, this MR scanner system can comprise motor control unit (for example being used for moving conveyor belt 216) and the computing unit (Fig. 2 is not shown) with corresponding motor.In DE 102 02 986 A1 this MR scanner system is described, the content of this application is introduced in this with as a reference.
Fig. 3 illustrates illustrating of the original image 31,38 that comprises structural region 44, and this zone 44 comprises high-frequency signal.The zone 44 that comprises high-frequency signal can be at the edge between structural area on the high level 42 and the structural area 43 on low-level.Reference numeral 32,45 expressions are positioned at two pixels on this edge or the step 44.
As can seeing from the image of being played up 39, aliasing causes the staircase artifacts at the edge 44 in institute's rendering image 39, and this is because the sample frequency that is sampled in the image 33 is not high enough.This causes deterioration of image, when with high frame rate render view or when generating film (in many virtual endoscopies are used, being exactly situation for example), and picture quality even can become poorer.Because pixel value switches between different gray-scale values at once when the edge exceeds pixel center, thereby make the edge that passes image pixel produce flicker.This typical aliasing effect is visually horrible and should be eliminated, so that reach the target of excellent picture quality.
The over-extraction sample is a kind of method of removing the aliasing pseudomorphism, as can be from being sampled picture 34 and being seen the rendered picture 40 accordingly.Here, be not only to adopt a sample point for each pixel, but assess 16 or more multisample point value and give this pixel with their mean value.Can obtain level and smooth edge like this, but also the flicker in the film is disappeared, this is that pixel value gradually changes between gray level because when passing the edge.Yet this method assesses the cost higher and to play up the target of performance inconsistent with height.Even in the time only need carrying out the over-extraction sample to the several pixels in the image, it also is unacceptable playing up slowing down of speed.
Therefore, according to an one exemplary embodiment of the present invention, by carry out the zone that comprises high-frequency signal (being the edge 44 among Fig. 3) that described structure is discerned in the pre-sampling or second sampling with the pre-sampling rate or second sampling rate.This pre-sampling rate is lower sampling rate, thereby obtains quick sampling, but still has sufficiently high sampling rate for the structural region that does not comprise high-frequency signal.After identifying the zone that comprises high-frequency signal, along traversing this regional track 36 sampling rate or first sampling rate execution post-sampling or first sampling later on that comprises high-frequency signal.In addition, carry out another post-sampling along another track 37 that traverses this zone that comprises high-frequency signal.This post-sampling obtains comprising the structure that is sampled of first sample point 46,47.
According to an one exemplary embodiment of the present invention, this track can be a line (as shown in Figure 3) that traverses the described zone that comprises high-frequency signal.But should be noted that this track can have any other shape, for example convex or spill.For example, this track can be arranged along the point on the surperficial change direction of maximum.
In other words, according to an aspect of the present invention, select sample point along the direction of maximum surface normal variation.Therefore, only need just can be with suitable quality rendering image than other over-extraction sample or anti-aliasing technology appended sample still less.Thus, anti-aliasingly on calculating, become feasible, even for also being like this play up aspect of performance the application of high request being arranged.
Select the sample point 46,47 that adds adaptively based on edge orientation, as can from being sampled of Fig. 3 see the image 35.Because surface normal variation mainly appears on the direction perpendicular to the edge, surface normal can not have significant change on parallel direction simultaneously, therefore assess appended sample seldom, and adopt mean value (for example weighted mean value) as pixel value along vertical direction.Can project to by the surface normal that will have pixel (pixel 49) and determine this vertical direction on the plane of delineation than the low depth value.Be enough to image quality rendering image along the additional sample points seldom of this direction, as can be from being seen the rendering image 41 with brilliance.
It should be noted that, can determine described structure by isodensity surface rendering program, itself in addition can carry out interactive threshold value adjustment, thereby (onthe fly) finds the possibility of appropriate threshold and the possibility of checking interesting feature in certain threshold range for interactive threshold value is determined to provide in operation.
Fig. 4 illustrates employing 401 and does not adopt 402 according to the anti-aliasing virtual coloscope check image of playing up of the self-adaptation that comprises the over-extraction sample of an one exemplary embodiment of the present invention.If do not satisfy Nyquist-Shannon sampling law,, then do not adopt the high-frequency signal of the image that self-adaptation anti-aliasing 401 plays up correctly to rebuild according to being sampled data if promptly sample frequency is lower than Nyquist rate.Because the unlimited frequency spectrum of these characteristics of image, this be aliasing in through one in the image of isodensity surface rendering obviously performance be staircase artifacts or along " zigzag pattern (jaggie) " at edge.In the motion picture or image played up with high frame rate, the defective of eye impressions becomes more obvious, and wherein the aliasing pseudomorphism causes the edge that glimmers, and neuroticism dispersion attention factor is introduced at described flicker edge.A clinical practice example that generates film is the virtual coloscope inspection.This tedious flicker makes the observer be difficult to be primarily focused on the actual task of seeking polyp, and therefore, the measure of this situation of any improvement all will receive an acclaim in clinical practice.
In image 401, can observe the aliasing pseudomorphism along the edge.Can significantly reduce the aliasing pseudomorphism for the additional ray of each pixel projection.Yet it is inconsistent that the additional calculations cost that is come by traditional over-extraction belt transect and height are played up performance target mutually.The present invention is by for example selecting additional sample points seldom to solve this problem perpendicular to the direction at edge.Play up data from the centre that the pre-sampling carried out with the pre-sampling rate or second sampling rate or second sampling draw and be used to determine these additional sample points.Experiment shows that 4 adaptively selected sample points of every pixel provide and 16 times of visual qualities that the over-extraction sample is suitable, but but has much lower assessing the cost, as can the image 402 from Fig. 4 seeing.
Fig. 5 illustrates the image of playing up with normal low sampling rate that demonstrates staircase artifacts 501, and wherein image 502,503 demonstrates anti-aliasing beneficial effect.When generating image with high frame rate, the edge seems smoothly and not to glimmer.The anti-aliasing quality of self-adaptation for 16 times of over-extraction sampled images of isodensity surface rendering introducing, but much lower assessing the cost but had.This can be from Figure 50 2 and sees Figure 50 3 by comparison, Figure 50 2 illustrates the over-extraction sampled images that adopts 16 samples for every pixel, only adopts 4 additional sample points and the self-adaptation according to an one exemplary embodiment of the present invention of Figure 50 3 is anti-aliasing for every pixel.
Fig. 6 illustrates the process flow diagram that the structure in the data set is carried out an one exemplary embodiment of adaptively sampled method according to the present invention.This method begins at step S0 place, afterwards, carry out data centralized procurement collection at step S1, this for example is by the polychromatic electromagnetic radiation source that produces the polychrome beam and (this is to check or the situation in the magnetic resonance virtual endoscopy for example at for example CT virtual coloscope) of being undertaken by the ray detector that detects this polychrome beam.
Afterwards, in step S2, carry out the pre-sampling or second sampling.This pre-sampling can be carried out with low sampling rate, this low sampling rate is for enough height for high image quality is provided in the lower frequency region, still for then not high enough for (such as the edge) in comprising the zone of high-frequency signal provides the picture quality of brilliance.After pre-sampling, carry out the isodensity surface rendering at step S3, so that by the structure in definite this data set of ray projection.Can carry out the ray projection to each pixel in the plane of delineation, so that calculate distance and the angle the surface normal of observing vector and this point from observation point to the respective table millet cake.Now, those image areas of the remarkable loss that can utilize described intermediate data to determine wherein to owe to sample will to cause picture quality are such as having zone or edge (for example surface ruffle) that high surface orientation changes.
In step S3, determine after the described structure zone that comprises high-frequency signal in step S4 in this structure of identification, promptly high surface orientation variation zone.Then, determine the direction perpendicular to this edge in step 5, this for example is to project on the plane of delineation by the surface normal that will have the pixel (seeing the Reference numeral 49 among Fig. 3) than the low depth value to carry out.Be enough to image quality rendering image along the additional sample points seldom of this direction with brilliance.
Then, in step S6, with as first sampling rate of high sampling rate along the line that traverses described edge, along carry out the post-sampling or first sampling perpendicular to the direction at this edge.In step S7, determine average pixel value along the weighted mean value of the sample point value of this line by assessment.Work can draw clearly the pixel value of determining like this, thereby and with the image quality of the low brilliance that obtains whole image of assessing the cost (even also be like this in edge).This method finishes at step S8 place.
It should be noted that, when playing up network model, the self-adaptation aliasing can also be used to eliminate the aliasing pseudomorphism along adjacent polygonal border, just as the situation in most of hardware-accelerated surface rendering programs, for example in all modification of " stepping cube (marching cube) " algorithm or in current three-dimensional computations machine game, be exactly this situation.In addition, so that glittering transparent surface when visual, described self-adaptation anti-aliasing method can expand to direct volume and play up whenever adopting shadow model (shading model) at the precipitous part place of transition function.
It is the Another application field that two dimensional image is handled, and wherein anti-aliasingly fast may prove useful.
Fig. 7 illustrates an one exemplary embodiment according to image processing equipment of the present invention of an one exemplary embodiment that is used to carry out the method according to this invention.Image processing equipment shown in Figure 7 comprises CPU (central processing unit) (CPU) or image processor 151, and it is connected to and is used for the storer 152 that the image of objects (such as the patient) is described in storage.Image processor 151 can be connected to a plurality of I/O networks or diagnostic device (such as MR equipment or CT equipment).This image processor also can be connected to display device 154 (for example computer monitor), calculates and adaptive information or image to be used for being presented at image processor 151.The operator can carry out reciprocation by keyboard 155 and/or other output device and image processor 151 not shown in Figure 7.
In addition, by bus system 153, Flame Image Process and processor controls 151 can also be connected to for example movement monitor, this movement monitor monitors the motion of objects.Carry out under the situation of imaging in for example lung to the patient, motion sensor can be an exhalation sensor.In the situation that heart is carried out imaging, motion sensor can be electrocardiograph (ECG).
Claims (13)
1, a kind of structure in the data set is carried out adaptively sampled method, this method may further comprise the steps: the zone that comprises high-frequency signal of discerning this structure; Carry out first sampling with first sampling rate, thereby draw the structure that is sampled that comprises each first sample point; Wherein, carry out this first sampling along traversing this regional track that comprises high-frequency signal.
2, method according to claim 1 wherein, is discerned the described structural region that comprises high-frequency signal by carry out second sampling with second sampling rate; And wherein, first sampling rate is higher than second sampling rate.
3, method according to claim 1 wherein, is played up the described structure that is sampled based on first sampling; And wherein, determine to comprise the average pixel value of the pixel in the described zone of high-frequency signal by the mean value of each first sample point value of assessment.
4, method according to claim 1 wherein, is determined described structure by isodensity surface rendering program.
5, method according to claim 1 wherein, is the line that traverses the described zone that comprises high-frequency signal along its described track of carrying out first sampling.
6, method according to claim 1 wherein, comprises that the described structural region of high-frequency signal is the edge.
7, method according to claim 1, wherein, the described adaptively sampled ray projection that comprises; And wherein, this ray projection is used to detect described edge.
8, method according to claim 1, wherein, described structure comprises the surface; Wherein, this surface comprises that the surface changes; And wherein, described track extends along the surperficial change direction of maximum.
9, method according to claim 1, wherein, this method is used to virtual endoscopy.
10, a kind of being used for carried out adaptively sampled image processing equipment to the structure in the data set, and this image processing equipment comprises storer and the image processor that is used to store this data set, and this image processor is suitable for carrying out following operation: load this data set; Discern the zone that comprises high-frequency signal of this structure; Carry out first sampling with first sampling rate, thereby draw the structure that is sampled that comprises each first sample point; Wherein, carry out first sampling along traversing this regional track that comprises high-frequency signal.
11, a kind of scanner system comprises the storer that is used for stored data sets and is suitable for the structure of this data centralization is carried out adaptively sampled image processor that wherein, this image processor is suitable for carrying out following operation: load this data set; Discern the zone that comprises high-frequency signal of this structure; Carry out first sampling with first sampling rate, thereby draw the structure that is sampled that comprises each first sample point; Wherein, carry out first sampling along traversing this regional track that comprises high-frequency signal.
12, scanner system according to claim 11, wherein, this scanner system is one of them of CT scan device system and MR scanner system.
13, a kind of being used for carried out adaptively sampled computer program to the structure in the data set, and wherein when carrying out this computer program on image processor, this computer program makes this image processor carry out following operation: load this data set; Discern the zone that comprises high-frequency signal of this structure; Carry out first sampling with first sampling rate, thereby draw the structure that is sampled that comprises each first sample point; Wherein, carry out first sampling along traversing this regional track that comprises high-frequency signal.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| EP04101030 | 2004-03-12 | ||
| EP04101030.7 | 2004-03-12 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1930588A true CN1930588A (en) | 2007-03-14 |
Family
ID=34960838
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2005800079332A Pending CN1930588A (en) | 2004-03-12 | 2005-03-03 | Adaptive sampling along edges for surface rendering |
Country Status (5)
| Country | Link |
|---|---|
| US (1) | US7742631B2 (en) |
| EP (1) | EP1728217A1 (en) |
| JP (1) | JP2007528769A (en) |
| CN (1) | CN1930588A (en) |
| WO (1) | WO2005091227A1 (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101673409B (en) * | 2009-09-11 | 2011-09-21 | 广州华多网络科技有限公司 | Image rendering method applied to computer screen |
| CN103310447A (en) * | 2013-06-12 | 2013-09-18 | 中国科学院光电技术研究所 | Two-dimensional plane sampling method based on self-adaptive technology |
| CN105225264A (en) * | 2014-06-27 | 2016-01-06 | 三星电子株式会社 | Based drive Adapti ve rendering |
| CN106725415A (en) * | 2016-11-15 | 2017-05-31 | 广州视源电子科技股份有限公司 | Method and device for processing electrophysiological signals |
| CN107110946A (en) * | 2014-09-12 | 2017-08-29 | Ge医疗系统环球技术有限公司 | Magnetic resonance device and program |
Families Citing this family (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7963695B2 (en) | 2002-07-23 | 2011-06-21 | Rapiscan Systems, Inc. | Rotatable boom cargo scanning system |
| US8275091B2 (en) | 2002-07-23 | 2012-09-25 | Rapiscan Systems, Inc. | Compact mobile cargo scanning system |
| US6928141B2 (en) | 2003-06-20 | 2005-08-09 | Rapiscan, Inc. | Relocatable X-ray imaging system and method for inspecting commercial vehicles and cargo containers |
| GB0426170D0 (en) | 2004-11-29 | 2004-12-29 | Falanx Microsystems As | Processing of computer graphics |
| US7471764B2 (en) | 2005-04-15 | 2008-12-30 | Rapiscan Security Products, Inc. | X-ray imaging system having improved weather resistance |
| CN101292260B (en) * | 2005-10-21 | 2014-12-10 | 皇家飞利浦电子股份有限公司 | Rendering method and apparatus |
| US20070297560A1 (en) * | 2006-03-03 | 2007-12-27 | Telesecurity Sciences, Inc. | Method and system for electronic unpacking of baggage and cargo |
| US8044971B2 (en) | 2008-01-31 | 2011-10-25 | Arm Norway As | Methods of and apparatus for processing computer graphics |
| US8115783B2 (en) | 2008-01-31 | 2012-02-14 | Arm Norway As | Methods of and apparatus for processing computer graphics |
| GB0803644D0 (en) | 2008-02-28 | 2008-04-02 | Rapiscan Security Products Inc | Scanning systems |
| GB0803641D0 (en) | 2008-02-28 | 2008-04-02 | Rapiscan Security Products Inc | Scanning systems |
| GB0809110D0 (en) | 2008-05-20 | 2008-06-25 | Rapiscan Security Products Inc | Gantry scanner systems |
| US20090322781A1 (en) * | 2008-06-30 | 2009-12-31 | Mikhail Letavin | Anti-aliasing techniques for image processing |
| US20100013854A1 (en) * | 2008-07-18 | 2010-01-21 | Microsoft Corporation | Gpu bezier path rasterization |
| KR101331047B1 (en) * | 2009-09-01 | 2013-11-19 | 한국전자통신연구원 | Rendering system of hair image and rendering method thereof |
| US9218933B2 (en) | 2011-06-09 | 2015-12-22 | Rapidscan Systems, Inc. | Low-dose radiographic imaging system |
| EP2952068B1 (en) | 2013-01-31 | 2020-12-30 | Rapiscan Systems, Inc. | Portable security inspection system |
| KR102370617B1 (en) | 2015-04-23 | 2022-03-04 | 삼성전자주식회사 | Method and apparatus for processing a image by performing adaptive sampling |
| WO2017201494A1 (en) * | 2016-05-19 | 2017-11-23 | Avantis Medical Systems, Inc. | Methods for polyp detection |
| CN114727791A (en) * | 2019-10-11 | 2022-07-08 | 得克萨斯系统大学评议会 | Variable zoom X-ray computed tomography method for composite materials |
| CN113541852B (en) | 2020-03-31 | 2022-06-10 | 华为技术有限公司 | Signal decoding method, decoding circuit and stylus pen |
| CN114061486B (en) * | 2021-11-19 | 2022-08-16 | 南京航空航天大学 | Automatic measuring device and method for large-scale skin curved surface of airplane |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2988691B2 (en) * | 1990-06-15 | 1999-12-13 | 株式会社東芝 | Image processing device |
| US6018600A (en) * | 1998-04-15 | 2000-01-25 | Arch Development Corp. | Sampling and reconstruction of signals and images including MR images of multiple regions |
| US6700672B1 (en) | 1999-07-30 | 2004-03-02 | Mitsubishi Electric Research Labs, Inc. | Anti-aliasing with line samples |
| DE10150428A1 (en) * | 2001-10-11 | 2003-04-30 | Siemens Ag | Multiple three dimensional imaging system for research integrates high and low resolution |
| DE10202986A1 (en) | 2002-01-26 | 2003-07-31 | Philips Intellectual Property | Coil system for an MR apparatus and MR apparatus with such a coil system |
| US6943805B2 (en) * | 2002-06-28 | 2005-09-13 | Microsoft Corporation | Systems and methods for providing image rendering using variable rate source sampling |
-
2005
- 2005-03-03 CN CNA2005800079332A patent/CN1930588A/en active Pending
- 2005-03-03 EP EP05708927A patent/EP1728217A1/en not_active Withdrawn
- 2005-03-03 WO PCT/IB2005/050791 patent/WO2005091227A1/en not_active Application Discontinuation
- 2005-03-03 JP JP2007502475A patent/JP2007528769A/en active Pending
- 2005-03-03 US US10/598,635 patent/US7742631B2/en not_active Expired - Fee Related
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101673409B (en) * | 2009-09-11 | 2011-09-21 | 广州华多网络科技有限公司 | Image rendering method applied to computer screen |
| CN103310447A (en) * | 2013-06-12 | 2013-09-18 | 中国科学院光电技术研究所 | Two-dimensional plane sampling method based on self-adaptive technology |
| CN103310447B (en) * | 2013-06-12 | 2016-01-13 | 中国科学院光电技术研究所 | Two-dimensional plane sampling method based on self-adaptive technology |
| CN105225264A (en) * | 2014-06-27 | 2016-01-06 | 三星电子株式会社 | Based drive Adapti ve rendering |
| CN107110946A (en) * | 2014-09-12 | 2017-08-29 | Ge医疗系统环球技术有限公司 | Magnetic resonance device and program |
| CN107110946B (en) * | 2014-09-12 | 2020-05-15 | Ge医疗系统环球技术有限公司 | Magnetic resonance apparatus and program |
| CN106725415A (en) * | 2016-11-15 | 2017-05-31 | 广州视源电子科技股份有限公司 | Method and device for processing electrophysiological signals |
Also Published As
| Publication number | Publication date |
|---|---|
| JP2007528769A (en) | 2007-10-18 |
| US20070177005A1 (en) | 2007-08-02 |
| EP1728217A1 (en) | 2006-12-06 |
| WO2005091227A1 (en) | 2005-09-29 |
| US7742631B2 (en) | 2010-06-22 |
| WO2005091227A8 (en) | 2006-10-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1930588A (en) | Adaptive sampling along edges for surface rendering | |
| US7853058B2 (en) | Determining a viewpoint for navigating a virtual camera through a biological object with a lumen | |
| US7116810B2 (en) | Method and system for airway measurement | |
| CN1930587B (en) | Image visualization | |
| US7447535B2 (en) | Mapping the coronary arteries on a sphere | |
| Summers et al. | Virtual bronchoscopy: segmentation method for real-time display. | |
| US8081809B2 (en) | Methods and systems for optimizing high resolution image reconstruction | |
| US8229200B2 (en) | Methods and systems for monitoring tumor burden | |
| CN109544534A (en) | A kind of lesion image detection device, method and computer readable storage medium | |
| US9449381B2 (en) | Methods, systems, and media for generating and analyzing medical images having elongated structures | |
| US9858716B2 (en) | Fast three-dimensional visualization of object volumes without image reconstruction by direct display of acquired sensor data | |
| JP2004105731A (en) | Processing of computer aided medical image | |
| CN1502310A (en) | Method and system for measuring disease relevant tissue changes | |
| US20040202990A1 (en) | System and method for performing a virtual endoscopy | |
| EP1861825A1 (en) | Method and device for the iterative reconstruction of cardiac images | |
| CN108269292B (en) | Method and device for generating two-dimensional projection images from three-dimensional image data sets | |
| CN1895185A (en) | Method for displaying information of checking region of checked object and influence of drug in vivo. | |
| CN1707523A (en) | Method for medical 3D image display and processing, computed tomograph, workstation and computer program product | |
| US20130223706A1 (en) | Quantification of a characteristic of a lumen of a tubular structure | |
| CN114565711A (en) | Heart image reconstruction method and system based on deep learning | |
| CN1531904A (en) | Method of operating imaging medical detection device | |
| CN1711562A (en) | Method for depicting an object displayed in a volume data set | |
| Zhou et al. | Estimating the center of rotation of tomographic imaging systems with a limited number of projections | |
| US20230048231A1 (en) | Method and systems for aliasing artifact reduction in computed tomography imaging | |
| Turlington et al. | Improved techniques for fast sliding thin-slab volume visualization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20070314 |